paper-with-me

홈 › Papers

Pixel-Level Cycle Association: A New Perspective for Domain Adaptive Semantic Segmentation

2020-10-31 · NeurIPS 2020 12 · Guoliang Kang, Yunchao Wei, Yi Yang, Yueting Zhuang, Alexander G. Hauptmann

Domain adaptive semantic segmentation aims to train a model performing satisfactory pixel-level predictions on the target with only out-of-domain (source) annotations. The conventional solution to this task is to minimize the discrepancy between source and target to enable effective knowledge transfer. Previous domain discrepancy minimization methods are mainly based on the adversarial training. They tend to consider the domain discrepancy globally, which ignore the pixel-wise relationships and are less discriminative. In this paper, we propose to build the pixel-level cycle association between source and target pixel pairs and contrastively strengthen their connections to diminish the domain gap and make the features more discriminative. To the best of our knowledge, this is a new perspective for tackling such a challenging task. Experiment results on two representative domain adaptation benchmarks, i.e. GTAV $\rightarrow$ Cityscapes and SYNTHIA $\rightarrow$ Cityscapes, verify the effectiveness of our proposed method and demonstrate that our method performs favorably against previous state-of-the-arts. Our method can be trained end-to-end in one stage and introduces no additional parameters, which is expected to serve as a general framework and help ease future research in domain adaptive semantic segmentation. Code is available at https://github.com/kgl-prml/Pixel- Level-Cycle-Association.

📄 PDF Abstract BibTeX arXiv:2011.00147

Code (1)

kgl-prml/Pixel-Level-Cycle-Association 공식 구현 pytorch

Tasks

Domain AdaptationSemantic SegmentationSynthetic-to-Real TranslationTransfer Learning

Similar Papers 제목 키워드 기반

CyCADA: Cycle-Consistent Adversarial Domain Adaptation

2017-11-08 · ICML 2018 7 · Judy Hoffman, Eric Tzeng, Taesung Park, Jun-Yan Zhu 외

Domain adaptation is critical for success in new, unseen environments. Adversarial adaptation models applied in feature spaces discover domain invariant representations, but are difficult to visualize and sometimes fail …

Domain AdaptationImage-to-Image TranslationSemantic SegmentationSynthetic-to-Real Translation+1

SPLAT: Semantic Pixel-Level Adaptation Transforms for Detection

2018-12-03 · Eric Tzeng, Kaylee Burns, Kate Saenko, Trevor Darrell

Domain adaptation of visual detectors is a critical challenge, yet existing methods have overlooked pixel appearance transformations, focusing instead on bootstrapping and/or domain confusion losses. We propose a Semanti…

Domain AdaptationPseudo LabelSemantic Segmentation

Learning Condensed and Aligned Features for Unsupervised Domain Adaptation Using Label Propagation

2019-03-12 · Jaeyoon Yoo, Changhwa Park, Yongjun Hong, Sungroh Yoon

Unsupervised domain adaptation aiming to learn a specific task for one domain using another domain data has emerged to address the labeling issue in supervised learning, especially because it is difficult to obtain massi…

Domain AdaptationUnsupervised Domain Adaptation

Multi-Camera Multi-Person Association using Transformer-Based Dense Pixel Correspondence Estimation and Detection-Based Masking

2024-08-17 · Daniel Kathein, Byron Hernandez, Henry Medeiros

Multi-camera Association (MCA) is the task of identifying objects and individuals across camera views and is an active research topic, given its numerous applications across robotics, surveillance, and agriculture. We in…

Dense Pixel Correspondence Estimation

Emotional Semantics-Preserved and Feature-Aligned CycleGAN for Visual Emotion Adaptation

2020-11-25 · Sicheng Zhao, Xuanbai Chen, Xiangyu Yue, Chuang Lin 외

Thanks to large-scale labeled training data, deep neural networks (DNNs) have obtained remarkable success in many vision and multimedia tasks. However, because of the presence of domain shift, the learned knowledge of th…

Domain AdaptationEmotion ClassificationEmotion RecognitionUnsupervised Domain Adaptation